Zitationsvorschlag
Lizenz (Kapitel)
Dieses Werk steht unter der Lizenz Creative Commons Namensnennung - Weitergabe unter gleichen Bedingungen 4.0 International.
Identifier (Buch)
Veröffentlicht
Classification of Genres through 500 Years of Spanish Literature in CORDE
Abstract In this work I analyze the development of numerous genres in almost five hundred years of Spanish literature. For that, I use a large diachronic corpus composed by the Real Academia Española. After an introductory section, the dataset is described, focusing on some important aspects of the distribution and balance of the categories. Next, several classification tests are applied in order to find out which parameters lead to the highest scores, and what the results for each category are. The variance of these results is then explored using linear regression, resulting in the length of the text being a good predictor for the classification results. Finally, the historical evolution is analyzed, showing that the classification results for genres neither get better nor worse over time, but remain stable.
Keywords: genre, classification, Spanish literature, corpora
